Nanofactory Proliferation

Because of the largely unexpected transformational power of molecular manufacturing, it is urgent to understand the issues raised. To date, there has not been anything approaching an adequate study of these issues.

CRN believes that at least thirty essential studies should be conducted immediately. Today we will examine study #22: "How can proliferation and use of nanofactories and their products be limited?"

Note: This is part of the third segment of studies, on "Policies and Policymaking". Recommended studies in this section assume the existence of a general-purpose molecular manufacturing system. All preliminary answers are based on diamondoid nanofactory technology.

This study will explore the challenge of preventing black markets, independent development, etc.

Subquestion A: How easy will it be to detect a development program?

Preliminary answer: Probably quite difficult. Development does not require exotic materials or massive industrial activity. It may require mainly off-the-shelf technology. Researchers will be from diverse and common fields like software engineering and computational chemistry, not concentrated in one exotic field. Depending on the bootstrapping 'recipe', the design effort might be dispersed (networked/teleconferenced), and the entire physical operation might be carried out in one moderate-sized laboratory. And most of the research would not require world-class talent, though a successful program today might well require world-class leadership.

Subquestion B: How much easier will it be to develop a second nanofactory, compared with developing the first one?

Preliminary answer: Reverse engineering will give hints as to which path to take. The definite knowledge that it can be done at all will reduce institutional friction. General technology advances will give a second program more to work with. Any leaks of know-how or software will further reduce the difficulty. It seems likely that the second nanofactory will be an order of magnitude less costly.

Subquestion C: How can nanoscale products be detected?

Preliminary answer: Unknown. Nanoporous filters can trap them. Non-proximal sub-wavelength optics, if they work as claimed, may be able to scan for them at a distance—but there are lots of natural nanoparticles, so recognition is also a problem. MRI may be able to detect at a distance, though resolution is a problem and there may be a theoretical limit.

Subquestion D: How easy will it be to smuggle nanofactories?

Preliminary answer: A fully functional nanofactory, able (given a supply of feedstock, energy, and blueprint software) to make one twice as big (and so on) and thus recreate a full manufacturing capacity, could be just a few microns on a side—small enough to hide inside a human cell. Or any convenient size in between. We don't know of any way to detect something like that without total intrusion of the volume being searched, which probably implies destruction.

Subquestion E: How easy will it be to detect proliferation-related activity?

Preliminary answer: Quite difficult. Especially once the 'recipe' is known, it will be very hard to spot a project—R&D for a nanofactory project may require only a single small lab and a few computers. (For comparison, consider Zyvex.)

Subquestion F: How effective will deterrence be?

Preliminary answer: To someone lacking a comparable capability, a nanofactory would be incredibly valuable. This implies that deterrence will not be successful.

Provisional conclusion: It will be very difficult to limit proliferation of nanofactory technology and possession of bootleg nanofactories.

Our initial basic findings (preliminary answers and provisional conclusions) for all thirty studies should be verified as rapidly as possible. Because our understanding points to a crisis, a parallel process of conducting these studies is strongly preferred.

We are actively looking for researchers who have an interest in performing or assisting with this work. Please contact CRN Research Director Chris Phoenix if you would like more information or if you have comments on the proposed studies.

Comments

[Tom Craver]

Chris: How about adding:

"G. Given the conclusion that proliferation will be difficult to prevent, how great is the danger?" To which the answer would be "Likely extreme. See other questions regarding nano-arms race, etc."

And then maybe "H. Is there then no hope?" with a provisional answer "Perhaps - try re-examining all premises behind answers A-G, to find an escape clause. We managed to avoid nuclear war through the MAD doctrine. Maybe if we start now, we can find a way."

[Tom Craver]

Following my own advice from above.

A premise behind the answer to "F" appears to be that the benefits of nanotechnology would not be available to someone who has not obtained the core capability of creating general purpose nanotechnology, capable of self-replication.

Therefore, what if nearly all the benefits of nanotechnology are made available to anyone who agrees not to seek development of self-replicating nanotechnology (SRnano); while simultaneously threatening extreme force (i.e. deterrence) against any who do seek that development?

This reduces the incremental value of obtaining or developing SRnano, without reducing the risks, hopefully tipping the answer to "F" toward successful deterrence.

As a side benefit, if an approach like this can be made acceptable to most of the world, any nation or group wishing to violate it would be greatly outnumbered, reducing the chance that they could use the element of surprise in an attack effective enough to avoid suffering the consequences of their actions.

The negative side, of course, is that this ultimately might require use of extreme measures to cure an "outbreak" of proliferation. So if a group of misguided idealists create a self-replicating "wish box" and start passing out copies to otherwise innocent friends and family, the enforcement could be ugly and controversial.

[Brett Bellmore]

Whatever policies might theoretically be implemented by people in such a position, we know only too well what sort of people gravitate to such positions: Power mad control freaks. Normal people aren't willing to engage in acts of extreme violence against people who are minding their own business.

The very measures that would be undertaken to maintain the monopoly, would provide more than ample justification for overthrowing it.

[Chris Phoenix, CRN]

Tom: *EXCELLENT* additions to this study. Thanks!

Your idea of making MNT benefits available to anyone who doesn't start their own program is quite reminiscent of our reasoning for advocating international development. This idea might in theory also be implementable if one nation develops far enough ahead of the others... though I find it hard to picture any nation scrapping their program and accepting restricted technology from another nation.

And after recent events, I'm finding it a bit harder to picture an international program successfully preempting or overshadowing nationalist programs.

Chris

[Tom Craver]

Chris: I agree that governments aren't going to be eager to give up on having their own SRnano. All will be concerned that they won't be able to defend themselves adequately. A few will be mainly concerned that they keep a lock on power.

Some nations can be trusted immediately - give them SRnano facilities of their own, provided only that they agree to a non-proliferation treaty. They'd have to agree to give SRnano only to nations approved by all other treaty members. Probably they'd also agree to enforce non-proliferation, if necessary.

There'll be nations that can't get unanimous support for treaty membership. Let any treaty member provide such a nation with all but SRnano, and provide SRnano services to that nation subject to design review by the treaty members.

Finally there'll be nations that no treaty member wants to give any kind of nanotech. Communicate directly with the people of that nation, letting them know that as soon as their government is gone or reformed, they can have the same benefits as the rest of the world is getting. Perhaps give them a taste of the benefits that will be available to them. Treaty members should monitor that nation, and be willing to squelch any MNT development effort, and go to war if they obtain MNT.

[michael vassar]

Since nanocomputers and neuro-scanning/reverse engineering appear to present the most serious MNT threat, and since there appears to be no realistic prospect of making this threat well understood, I find the idea of promoting any MNT development program to be very worrysome.

[Brett Bellmore]

I assume you're talking about what some call "uploading", Michael. How do you see this as a serious threat, as opposed to a great opportunity?

[Janessa Ravenwood]

Indeed. As someone who wants both a LOT of body alterations AND extensive memory/personality/mind alterations this research is critical for me.

[Chris Phoenix, CRN]

I think Michael is advancing the theory that any AI above a certain level is inherently dangerous, and the many-orders-of-magnitude improvement in computer power will make it easier to achieve that level. I'm not quite sure where brain scanning comes into it.

If this is a threat, it is not inherent in molecular manufacturing--computers are advancing by multiple orders of magnitude each decade anyway, and nanoscale technologies may speed that up. (I just read that Nantera has built a working 10 gigabit nanotube RAM using standard lithography technology.)

Chris

[Jim Logajan]

Change just one word of the question to yield: "How can proliferation and use of lathes and their products be limited?"

Lathes can be used to manufacture gun barrels. And cannon barrels. And are essential to the manufacture of a whole host of weapons - some quite horrific. But if you saw anyone propose the above, they'd be denounced as patronizing and elitist. They'd be denounced as fear-mongering. They'd be denounced for the assumptions made in the question (i.e. it presumes the answer to "Should the use of lathes and their products be limited?" is "yes").

So why is this question even being posed? Why is such a leading question called a "study"?

[Brett Bellmore]

I think where brain scanning comes into it, is that it at least theoretically allows you to build human level AIs without having to have a deep understanding of how intelligence works. You just need a lower level understanding of how neurons work, and the net listing of a working brain, and a lot of computing power, and "presto!", you can churn out black boxes which, even though you don't understand them at a high level of abstraction, are intelligent. And already educated, too.

Since it seems likely that the limiting factor after MNT is available is intellectual, being able to mass produce millions of copies of design engineers, possibly running at a much higher "clock speed" than human, would give anybody who did it a real boost over the competition.

And there ARE people who would volunteer for that, you know. Especially if you offered to include a really nice VR enviroment for them to work "in".

[Malcolm McCauley]

I believe the question is an attempt to see if such a method would be possible, at all. It's a good question to pose, when trying to make a policy decision; as many possibilities as possible should be explored...

I also believe that proliferation should be limited. Not to stop people from getting the technology; that will happen sooner or later. But proliferation should be controlled, so that non-destructive forces get ahold of the technology first. Which is what Tom had an idea on.

I don't think there needs to be any limitation of freedom, or even a limit to access to the technology. The flow of information and expertise has generally flowed in one direction, toward the people who try to build up the world, rather than destroying it. I think what should be done is to encourage this as much as possible; the more repressive forces are slower, they don't need any artificial hinderance.

I think Tom's idea is a good one; with a few adjustments, I think it would work. Take his idea, except don't try to stop the flow of information; expand it for the right people/places.

[Tom Craver]

Jim: I think the significant difference between a lathe and a nanofactory is that the former required a lot of other things to make it dangerous - skilled operators, skilled metal casters, etc. Only a fair sized government could really afford to the first guns. And even if anyone could use a lathe to easily make guns, at best they could kill a few handfuls of other people before being killed themself.

Now maybe the same will be true of early nanofactories - though most people seem to think that we'll be able to very quickly move from primitive nanotech to the desktop "wish box" nanofactory. And then any previously designed dangerous thing - from anthrax spores to pocket-sized laser-ignited H-bombs to Gray Goo - might be made by anyone smart enough to browse the web and click a few buttons on their computer screen.

I'm not opposed to the idea that somehow we should get to a world where everyone has free access to all of MNT's benefits - I'm just trying to figure out how we can survive in the face of the fact that there're bound to be a few nutballs who'll deliberately set out to wipe out whole cities, nations - or the world.

[Chris Phoenix, CRN]

Jim: your comparison is unfair. Lathes are a very limited technology. They can't make the firing mechanism of a gun. And they require substantial skill to operate. And they certainly aren't the easiest way to acquire guns.

No one has shown a meaningful limitation on what weapons a diamondoid nanofactory could produce. And it would be trivial to use. Product design may or may not be easy for advanced products; but for simple products, it certainly could be. It could be substantially easier to design and build a bomb with a nanofactory than with, say, a propane canister.

Tom, I'm not going as far as postulating a "wish box" that could make anthrax spores just for the asking. There would be limitations on the chemistry. And you couldn't wish--you'd have to design, or download a preexisting design. But if unrestricted nanofactories are available to everyone, we'd better hope that defense is vastly easier than offense--something that is far from clear.

Chris

[michael vassar]

This is NOT what I think brainscanning will lead to, but it should at least give enthusiasts pause
http://hanson.gmu.edu/uploads.html

This is a bit more speculative and complicated, possibility, but both more plausible and worse
http://www.nickbostrom.com/fut/evolution.html

Actually, I expect lots of smileys
see www.sl4.org/archive search for "smiley" in topics from 2004.

If you try to optimize anything, you use a really powerful technique for doing so, and you don't understand the thing you are optimizing, you will probably die.

[Janessa Ravenwood]

Mike V: read the links, don't fully agree with everything, but have no plans to upload myself and I seriously doubt the majority of people on the planet will be stampeding to do this anyway. I personally think you've got have a few screws loose to want to live as a disembodied copy of a person, but there are some people - not hundreds of millions mind you - that think this is a good idea for some reason. Oh well, I wouldn't worry about this becoming a major trend. The real worry is using knowledge of the mind for "non-elective" mind alterations to fix "incorrect beliefs" - Chinese anyone?

[Tom Craver]

Just an example of how the "real world" reacts to a realistic prospect of dangerous proliferation:

"What Israel fears is an Iranian nuclear weapon being given to terrorists, who would then smuggle it into Israel and detonate the "anonymous" nuke. This would be difficult, as Israeli border and port controls are strict. But it is the kind of terrorist nightmare that Israel has to deal with, and a bombing raid against Iran seems the lesser of two evils."

http://www.strategypage.com//fyeo/qndguide/default.asp?target=IRAN.HTM

Project that onto a future situation involving nanotech...

[Tom Craver]

Janessa - I agree that "non-elective" mind changes are a great danger. But I wouldn't dismiss the possibility of uploads being a major trend, if/when feasible.

Suppose a few wealthy businessmen on the verge of dying upload, and have just a few small advantages - like not getting tired or needing to sleep, and being able to split their attention a dozen different ways at once. Soon aging but otherwise healthy CEOs would be uploading, just to stay competitive.

With the stigma of "being software" removed, upper-management CEO-wanna-bees jump on the "software management" bandwagon. They're soon followed by their subordinates, who can clearly see the road to promotion lies in going to bits.

And that's just one path to greatly increased acceptance, and assumes very modest benefits from uploading.

[Mike Deering]

"What happens when almost anything you wish for can suddenly pop into existence, at your command?"

Certainly, producing living organisms such as sex partners is within the theoretic capability of end-stage molecular manufacturing, but, the first generation of nanofactory will be constrained by several factors:

First and foremost will be simplicity of design of the nanofactory core technology.   Wet biological nanotech is very complicated because of all the different kinds of chemical bonds involved and difficult to control precisely because it is rather delicate.   Vacuum diamondoid, fullerene, and carbon nano-tube (CNT) assembly is much simpler by comparison, almost all carbon-carbon bonds, which are the strongest bonds in chemistry, resulting in extreme stability.   ( note: diamondoid nanotechnology is understood to include Lomer dislocation non-cleaving plane synthetic diamond, fullerenes and CNTs.)   Consequently, the first nanofactories will produce solely inanimate, near chemically inert, almost indestructible products.   These products will have capabilities which include structural, electronic, mechanical, optical, and computational capability.   But don't think that you can't have soft textures and flexible materials with diamondoid construction. Nanotubes are flexible.   A properly designed assemblage of CNTs can achieve any macroscopic physical characteristics desired, while maintaining near indestructibility.   Producing the sex slave of your dreams solely comprised of diamondoid nanotechnology, indestiquishable from the biological analog, is merely a design challenge, not a technology limitation.   On the other hand, a diamondoid technology nanofactory can't produce a ham sandwich.

The second constraint will be security.   The restriction against producing nanobots that self replicate without appropriate limit in any natural environment, atmospheric, terrestrial, aquatic, subterranean, biological, or outer space.   This restriction can be accomplished by limiting the nanofactory to producing products solely constructed from approved safe nanoblocks.   As states above, these nanoblocks would have capabilities that include structural, electronic, mechanical, optical, and computational.   The missing capability from these nanoblocks will be molecular assembly.  The list of things that could not be produced from these nanoblocks will include: chemical and nuclear explosives, biological agents (viruses, bacteria), living organisms, in short, nuclear/biological/chemical (NBC) weapons.   Of course, some of these weapons can be produced with the help of diamondoid nanofactory produced equipment.   This subsidiary threat can only be managed with total nanotech restriction (unacceptable) or traditional anti-criminal/anti-terrorist methods, intelligence, surveillance, and restriction of proven bad actors.

The third constraint will be sociological.   The restriction on producing products determined to be unhealthy for society, drugs, wireheading devices, sex partners, pornography, basically anything that could be any fun.   This is a matter that society will have to work out the limits of for itself.   My personal opinion is that no sociological restrictions are justified.   This restriction will have to be implemented by internal nanofactory AGI or a system or pre-approval authorization for product files.

The fourth constraint will be intellectual property (IP).   This involves copyrights, patents, usage licensing, financial reimbursement of design costs, and other economic considerations (corporate greed).   Not that I have anything against greed.   Greed is good.   Greed works.   At least it works in an environment of scare or limited resources.   My opinion is that it is a negative factor in an abundance economy.   This would likewise be regulated with some sort of copy protection or authorization system legal or technological.

[Brett Bellmore]

"What happens when almost anything you wish for can suddenly pop into existence, at your command?"

They covered THAT scenario in "Forbidden Planet". LOL